Kernel Interpolation for Scalable Online Gaussian Processes
Kernel Interpolation for Scalable Online Gaussian Processes
复制标题
DOI：
--
复制
发表时间：
2021-03
影响因子：
2.7
通讯作者：
S. Stanton;Wesley J. Maddox;Ian A. Delbridge;A. Wilson
中科院分区：
文献类型：
--
作者：
S. Stanton;Wesley J. Maddox;Ian A. Delbridge;A. Wilson
高斯过程（GP）为在线设置中的性能提供了黄金标准，例如样本有效控制和黑盒优化，我们需要在连续采集数据时更新后验分布。然而，更新GP后，以适应即使是一个单一的新的观察后，观察到$n$点，在确切的设置至少会导致$O（n）$的计算。我们展示了如何使用结构化内核插值来有效地回收计算的恒定时间O（1）$在线更新的点$n$的数量，同时保留精确的推理。我们展示了我们的方法在一系列在线回归和分类设置，贝叶斯优化和主动采样，以减少疟疾发病率预测的错误的承诺。代码可在https://github.com/wjmaddox/online_gp上获得。
Gaussian processes (GPs) provide a gold standard for performance in online settings, such as sample-efficient control and black box optimization, where we need to update a posterior distribution as we acquire data in a sequential fashion. However, updating a GP posterior to accommodate even a single new observation after having observed $n$ points incurs at least $O(n)$ computations in the exact setting. We show how to use structured kernel interpolation to efficiently recycle computations for constant-time $O(1)$ online updates with respect to the number of points $n$, while retaining exact inference. We demonstrate the promise of our approach in a range of online regression and classification settings, Bayesian optimization, and active sampling to reduce error in malaria incidence forecasting. Code is available at https://github.com/wjmaddox/online_gp.